Approximately 8% of adults in the U.S. suffer from type 2 diabetes. Treatment prevents some of its devastating complications, but does not usually restore normal glycemia or eliminate adverse consequences. Hence, prevention is preferable. Evidence from two large, well-controlled trials indicates that type 2 diabetes can be delayed, and possibly prevented, through lifestyle interventions. Nutrition and exercise to achieve a healthy weight are key behavioral interventions in preventing diabetes. However, this evidence has not diffused well into clinical practice. Clinic-based weight management programs are expensive, difficult to deliver on a population basis and are not typically structured to meet the unique needs of medically underserved and Hispanic populations. (Prevalence of diabetes is especially high among Hispanic adults.) [unreadable] [unreadable] The goal of this proposal is to test the efficacy of a practical and easily delivered intervention for weight management in a balanced randomized controlled trial, involving 150 people in each of two study arms (control group and intervention group) for a total of 300 patients. This study will target medically underserved patients with a risk for diabetes. The primary outcomes will be to reduce their weight and improve glucose metabolism. The intervention will utilize an interactive computer-based software program to tailor the intervention to the unique needs of each patient. This system was developed under a Phase I SBIR award (1 R43- DK066711-01A1). Evidence-based behavioral models that hold promise for helping individuals make changes to better manage body weight will be applied. This system is designed to operate on clinic-based personal computers and will provide individual assessment and tailored, real-time and feedback reports to patients and physician reports outlining key areas for patient improvement and counseling suggestions for practitioners. [unreadable] [unreadable]